stylistic fixes, cleaned imports, fixed cyclic imports

nltk/align/ibm1.py

@@ -15,7 +15,6 @@
 from __future__  import division
 from collections import defaultdict
 from nltk.align  import AlignedSent
-from nltk.corpus import comtrans
 
 class IBMModel1(object):
     """
@@ -28,6 +27,7 @@ class IBMModel1(object):
     Step 2 - Estimate the probability of translation according to the 
              evidence from Step 1. 
 
+    >>> from nltk.corpus import comtrans
     >>> bitexts = comtrans.aligned_sents()[:100]
     >>> ibm = IBMModel1(bitexts, 20)
 

nltk/align/ibm2.py

@@ -9,7 +9,6 @@
 from __future__  import division
 from collections import defaultdict
 from nltk.align  import AlignedSent
-from nltk.corpus import comtrans
 from nltk.align.ibm1 import IBMModel1
 
 class IBMModel2(object):
@@ -26,6 +25,7 @@ class IBMModel2(object):
     Step 3 - Estimate the probability of translation and alignment according 
              to the evidence from Step 2. 
 
+    >>> from nltk.corpus import comtrans
     >>> bitexts = comtrans.aligned_sents()[:100]
     >>> ibm = IBMModel2(bitexts, 5)
     >>> aligned_sent = ibm.align(bitexts[0])

nltk/classify/tadm.py

@@ -14,7 +14,7 @@ from nltk.internals import find_binary
 try:
     import numpy
 except ImportError:
-    numpy = None
+    pass
 
 _tadm_bin = None
 def config_tadm(bin=None):

nltk/classify/textcat.py

@@ -31,14 +31,10 @@ http://borel.slu.edu/crubadan/index.html
 # Ensure that literal strings default to unicode rather than str.
 from __future__ import print_function, unicode_literals
 
-import nltk
-import nltk.compat
-from nltk.corpus import CrubadanCorpusReader
+from nltk.compat import PY3
 from nltk.util import trigrams
-from nltk.tokenize import word_tokenize
-from nltk.probability import FreqDist
 
-if nltk.compat.PY3:
+if PY3:
     from sys import maxsize
 else:
     from sys import maxint
@@ -71,7 +67,8 @@ class TextCat(object):
                                    "see https://pypi.python.org/pypi/regex for "
                                    "further details.")
 
-        self._corpus = CrubadanCorpusReader(nltk.data.find('corpora/crubadan'), '.*\.txt')
+        from nltk.corpus import crubadan
+        self._corpus = crubadan
         # Load all language ngrams into cache
         for lang in self._corpus.langs():
             self._corpus.lang_freq(lang)
@@ -82,6 +79,8 @@ class TextCat(object):
     
     def profile(self, text):
         ''' Create FreqDist of trigrams within text '''
+        from nltk import word_tokenize, FreqDist
+
         clean_text = self.remove_punctuation(text)
         tokens = word_tokenize(clean_text)
         
@@ -115,7 +114,7 @@ class TextCat(object):
             # Arbitrary but should be larger than
             # any possible trigram file length
             # in terms of total lines
-            if nltk.compat.PY3:
+            if PY3:
                 dist = maxsize
             else:
                 dist = maxint
@@ -148,41 +147,47 @@ class TextCat(object):
         return min(self.last_distances, key=self.last_distances.get)
         #################################################')
 
-    def demo(self):
-        from nltk.corpus import udhr
-
-        langs = ['Kurdish-UTF8', 'Abkhaz-UTF8', 'Farsi_Persian-UTF8',
-                 'Hindi-UTF8', 'Hawaiian-UTF8', 'Russian-UTF8', 'Vietnamese-UTF8',
-                 'Serbian_Srpski-UTF8','Esperanto-UTF8']
-
-        friendly = {'kmr':'Northern Kurdish',
-                    'abk':'Abkhazian',
-                    'pes':'Iranian Persian',
-                    'hin':'Hindi',
-                    'haw':'Hawaiian',
-                    'rus':'Russian',
-                    'vie':'Vietnamese',
-                    'srp':'Serbian',
-                    'epo':'Esperanto'}
+def demo():
+    from nltk.corpus import udhr
+
+    langs = ['Kurdish-UTF8', 'Abkhaz-UTF8', 'Farsi_Persian-UTF8',
+             'Hindi-UTF8', 'Hawaiian-UTF8', 'Russian-UTF8', 'Vietnamese-UTF8',
+             'Serbian_Srpski-UTF8','Esperanto-UTF8']
+
+    friendly = {'kmr':'Northern Kurdish',
+                'abk':'Abkhazian',
+                'pes':'Iranian Persian',
+                'hin':'Hindi',
+                'haw':'Hawaiian',
+                'rus':'Russian',
+                'vie':'Vietnamese',
+                'srp':'Serbian',
+                'epo':'Esperanto'}
         
-        for cur_lang in langs:
-            # Get raw data from UDHR corpus
-            raw_sentences = udhr.sents(cur_lang)
-            rows = len(raw_sentences) - 1
-            cols = list(map(len, raw_sentences))
+    tc = TextCat()
+
+    for cur_lang in langs:
+        # Get raw data from UDHR corpus
+        raw_sentences = udhr.sents(cur_lang)
+        rows = len(raw_sentences) - 1
+        cols = list(map(len, raw_sentences))
 
-            sample = ''
+        sample = ''
           
-            # Generate a sample text of the language
-            for i in range(0, rows):
-                cur_sent = ''
-                for j in range(0, cols[i]):
-                    cur_sent += ' ' + raw_sentences[i][j]
+        # Generate a sample text of the language
+        for i in range(0, rows):
+            cur_sent = ''
+            for j in range(0, cols[i]):
+                cur_sent += ' ' + raw_sentences[i][j]
             
-                sample += cur_sent
+            sample += cur_sent
           
-            # Try to detect what it is
-            print('Language snippet: ' + sample[0:140] + '...')
-            guess = self.guess_language(sample)
-            print('Language detection: %s (%s)' % (guess, friendly[guess]))
-            print('#' * 140)
+        # Try to detect what it is
+        print('Language snippet: ' + sample[0:140] + '...')
+        guess = tc.guess_language(sample)
+        print('Language detection: %s (%s)' % (guess, friendly[guess]))
+        print('#' * 140)
+
+
+if __name__ == '__main__':
+    demo()

nltk/corpus/reader/crubadan.py

@@ -22,7 +22,7 @@ http://borel.slu.edu/crubadan/index.html
 from __future__ import print_function, unicode_literals
 
 import re
-import nltk.compat
+from nltk.compat import PY3
 from os import path
 from nltk.corpus.reader import CorpusReader
 from nltk.probability import FreqDist
@@ -75,7 +75,7 @@ class CrubadanCorpusReader(CorpusReader):
         if self._LANG_MAPPER_FILE not in self.fileids():
             raise RuntimeError("Could not find language mapper file: " + mapper_file)
 
-        if nltk.compat.PY3:
+        if PY3:
             raw = open(mapper_file, 'r', encoding='utf-8').read().strip()
         else:
             raw = open(mapper_file, 'rU').read().decode('utf-8').strip()
@@ -93,13 +93,13 @@ class CrubadanCorpusReader(CorpusReader):
             raise Runtime("Could not find language n-gram file for " + lang)
 
         counts = FreqDist()
-        if nltk.compat.PY3:
+        if PY3:
             f = open(ngram_file, 'r', encoding='utf-8')
         else:
             f = open(ngram_file, 'rU')
 
         for line in f:
-            if nltk.compat.PY3:
+            if PY3:
                 data = line.split(' ')
             else:
                 data = line.decode('utf8').split(' ')

nltk/parse/malt.py

@@ -15,7 +15,6 @@ from functools import reduce
 import subprocess
 
 from nltk.data import ZipFilePathPointer
-from nltk.tag import RegexpTagger
 from nltk.tokenize import word_tokenize
 from nltk.internals import find_binary
 
@@ -43,6 +42,7 @@ class MaltParser(ParserI):
         if tagger is not None:
             self.tagger = tagger
         else:
+            from nltk.tag import RegexpTagger
             self.tagger = RegexpTagger(
             [(r'^-?[0-9]+(.[0-9]+)?$', 'CD'),   # cardinal numbers
              (r'(The|the|A|a|An|an)$', 'AT'),   # articles

nltk/parse/nonprojectivedependencyparser.py

@@ -14,7 +14,6 @@ import logging
 from nltk.compat import xrange
 
 from nltk.parse.dependencygraph import DependencyGraph
-from nltk.classify import NaiveBayesClassifier
 
 logger = logging.getLogger(__name__)
 
@@ -111,6 +110,8 @@ class NaiveBayesDependencyScorer(DependencyScorerI):
         :param graphs: A list of dependency graphs to train the scorer.
         """
 
+        from nltk.classify import NaiveBayesClassifier
+
         # Create training labeled training examples
         labeled_examples = []
         for graph in graphs:

nltk/parse/transitionparser.py

@@ -12,10 +12,14 @@ import pickle
 from os import remove
 from copy import deepcopy
 from operator import itemgetter
-from scipy import sparse
-from numpy import array
-from sklearn.datasets import load_svmlight_file
-from sklearn import svm
+
+try:
+    from numpy import array
+    from scipy import sparse
+    from sklearn.datasets import load_svmlight_file
+    from sklearn import svm
+except ImportError:
+    pass
 
 from nltk.parse import ParserI, DependencyGraph, DependencyEvaluator